Organizational Change “From Within”: Exploring the World Bank’s Early Lending Practices
Jeffrey M. Chwieroth
Department of International Relations
London School of Economics
Houghton Street
London WC2A 2AE
United Kingdom
Email: [email protected]
TEL: +44 (0) 20 7955 7209
FAX: +44 (0) 20 7955 7446
Draft
Abstract. The World Bank is delegated the authority to provide long-term stabilization and general
purpose balancing of payments loans, what the Bank later described as ―program loans.‖ Yet after a
brief period in its early years, the Bank stopped disbursing these loans for an extended period. Why?
I show that changes to the Bank‘s early lending practices can be understood largely as a product of
intra-organizational dynamics and change that facilitated the construction of a ―project-oriented
culture‖ that delegitimized the disbursement of program loans. Though providing a number of
compelling reasons to expect autonomous behaviour from international organizations (IOs), such
dynamics present a blind spot for rationalist approaches, which offer little insight into the processes that
shape preference formation ―from within‖ IOs. I augment constructivist approaches by going beyond
structuralist and static conceptions of IO staff as simply reacting to initiatives ―from above‖ or ―from
below‖ and offering a more dynamic explanation of organizational change that describes how internal
norm entrepreneurs emerge and proactively and strategically reconstruct an IO‘s culture. Emphasis is
placed on the role of personnel and internal institutional configurations.
Keywords: organizational culture, constructivism, World Bank, norm entrepreneurs, balance of
payments financing, institutions
1
What explains change in the behaviour of international organizations (IOs)? The traditional
view tends to be that external agents and states in particular tend to be responsible for IO change
―from above.‖ While realists emphasize demands imposed by powerful states, neoliberal
institutionalists often presume that IOs are generally responsive to such demands; otherwise, states
would presumably end their support. Alternatively, some scholars argue that non-governmental
organizations (NGOs) or trans-national advocacy networks can ―from below‖ to foment IO change.
Yet recent research from both sides of the rationalist and constructivist divide suggests that IOs can
develop considerable autonomy from their external environment. Recognizing that IOs can have
considerable autonomy suggests that we need to investigate and theorize the processes through
which IO change can come ―from within.‖
Because of the state-centric bias of international relations theory and the general
presumption that IOs have little autonomy, there have been few studies that investigate this
possibility. But change ―from within‖ IOs is a common, if relatively unexplored, phenomena.
Indeed, a growing number of scholars have begun to devote much needed attention to such
processes (Barnett and Coleman 2005; Chwieroth 2008). This article builds on these initial efforts,
exploring one little-studied but critical episode that decisively shaped the World Bank‘s early
lending practices.
The International Bank for Reconstruction and Development (IBRD) was designed at the
Bretton Woods Conference in 1944 to serve primarily as a supplier of long-term capital to promote
reconstruction and development. The IBRD was also delegated the authority to provide long-term
stabilization loans – loans with a long repayment term aimed at reconstituting the foreign reserves
of countries faced with long-term structural payments imbalances associated with post-war
reconstruction – and ―general purpose‖ balance of payments financing – loans with a shorter
repayment term that aim to finance payments imbalances but that fail to rigorously specify the use
of the proceeds. Though the IBRD never disbursed a long-term stabilization loan, in the immediate
post-war period it did extend a number of general purpose loans, which the Bank later identified as
2
―program loans‖ to differentiate them from ―project loans‖ where the proceeds are tied to specific
investments or imports.
But in 1957 the Bank stopped disbursing program loans for an extended period, even though
member states on the Executive Board had generally welcomed such loans enthusiastically. This
constitutes a puzzle for state-centric theories. Since programs loans offered the potential for larger
loans than projects loans, it is also rather perplexing for arguments that attribute IO behaviour solely
to budget maximization (Niskanen 1971). This remarkable case of an IO permitting one of its tasks
to atrophy also belies the predictions of those who assert that bureaucratic tendencies lead IOs to
engage in task expansion (Barnett and Finnemore 2004).
This article offers an alternative explanation that argues that the Bank‘s early lending
practices can be understood largely as a product of intra-organizational dynamics and change. State
of the art rationalist approaches that draw on principal-agent (PA) theory provide a number of
compelling reasons to expect autonomous IO behaviour. Yet while expecting organizational
change to often occur ―from within,‖ PA theorists offer little insight into the internal processes that
shape IO preference formation and change. Instead, PA theorists (Hawkins et al. 2006a:7) concede
that IO (agent) preferences ―are ‗filled in‘ as necessary by the specific assumptions of particular
theories,‖ even though these assumptions, as I show below, can often prove problematic.1
To be sure, factors highlighted by PA theorists did matter. These factors shaped the Bank‘s
authorizing environment, or what Woods (2006:4) calls the ―outer structural constraint,‖ within
which the management and staff operated. Ambiguous, conflicting, and complex mandates and
asymmetric information helped to create a permissive context in which the management and staff
could act autonomously. But PA theory cannot fully explain the evolution of the Bank‘s early
lending practices.
PA theory, as some leading proponents (Lake and McCubbins 2006:344)
acknowledge, is limited by ―its inability to make general claims about what agents do with the
autonomy they possess.‖
1
The inability of PA to offer general propositions about agent preferences is a recurring theme in PA analysis, see Lake
and McCubbins 2006:343n1, 344.
3
Why, then, did the Bank use its autonomy to cease disbursement of program loans? To
provide some answers to this question, I focus on internal collectively shared beliefs, what some
constructivists call organizational culture, that shape the Bank‘s staff‘s behaviour. These
collectively shared beliefs are critical for understanding staff behaviour as they specify legitimate
organizational goals and means.
Organizational change, I argue, often depends on norm entrepreneurs within an IO who
strategically promote greater acceptance of a particular practice or behaviour. Such entrepreneurs
come in different forms and play varying roles at different stages of the process, but existing studies
tend to focus narrowly on how change comes ―from above‖ or ―from below.‖ This neglect is
usually coupled with a tendency to downplay the possibility and efficacy of staff strategic agency.
The staff‘s main behavioural trait is typically one of reaction to external initiatives for change.
I therefore seek to inject greater dynamism and staff strategic agency into our understanding
of organizational change, showing how it can come ―from within.‖ Building on constructivist
approaches to IOs, this article seeks to identify the internal mechanisms through which norm
entrepreneurs surface within an IO and the conditions that facilitate their efforts. Emphasis is
placed on the role of personnel and internal institutional configurations.
Staff beliefs tend to be linked to the views that prevail within the profession(s) from which
an organization recruits. Professions are major agents in the construction of interpretations and
associated standards of behaviour. Recruits help bring these beliefs into an IO. Yet when an
organization, such as the Bank, recruits its staff from multiple professions, it can serve to facilitate
the development of subcultures, which will tend to reflect different professional beliefs.
These subcultures serve as a fertile source of norm entrepreneurs who promote change. The
development of various subcultures and the emergence of norm entrepreneurs often eventually
results in a ―battle of ideas‖ between proponents of different beliefs. In this battle, I argue the
discursive influence of norm entrepreneurs depends in large part on their position within the
organization. By highlighting these intra-organizational dynamics, this article brings into sharper
4
focus the decisive influence that the political battles within agents can have on IO behaviour as well
as the messiness of the delegation process in an IO‘s infancy. It also underscores the importance of
paying closer attention to how the staff shape organizational mandates and culture and to the
important, but understudied, role of IO presidents and managing directors.
The article is divided into five main sections. The first section outlines the PA approach.
The second section then evaluates the ability of this approach to explain the Bank‘s early lending
practices. Section three examines the constructivist approach and augments it with greater
attention to how change can occur ―from within.‖ The fourth section then applies this framework
for understanding the evolution of the Bank‘s early lending practices. The paper concludes with a
summary of the findings and a discussion of the key conclusions.
Principals, Agents, and Organizational Change
State of the art rationalist approaches treat IOs as agents and their member states as their
principals (Gould 2003; Hawkins et al. 2006b; Pollack 2003; Nielson and Tierney 2003). PA
theorists expect autonomous
PA theorists expect autonomous IO (agent) behaviour within ―zones
of discretion‖ that are based on member state (principal) calculations that include the benefits of
delegation and the costs of control mechanisms to rein IOs in. A number of conditions are said to
further IO autonomy and contribute to agent performance problems.
The act of delegation itself is often critical. Though necessarily involving a conditional
grant of authority, some acts of delegation are likely to prompt greater autonomy than others.
Ambiguous, conflicting, or complex mandates – what Gutner (2005) labels ―antinomic delegation‖
– can often contribute to agent performance problems. Gutner suggests that this type of delegation
frequently confronts multilateral development banks, which are typically torn between competing
mandates to serve as a ―bank‖ and a ―development agency.‖ Indeed, this problem surfaced almost
immediately after the IBRD was created. PA theorists also suggest that autonomous behaviour
arises from asymmetric information and collective action problems among and within member
states.
5
When IOs engage in behaviour that runs counter to the interests of member states, states are
expected to rein them using a variety of control mechanisms. These control mechanisms include
rewriting the organizational mandate, issuing operational directives, and ―screening and selecting‖
management and staff. These control mechanisms often come with normative strings attached to
them that can, over time, lead the staff to adopt new beliefs and change their behaviour. Though
PA theorists fully expect IOs to act on their own initiative, given their desire to elucidate agency
problems and the efficacy of various control mechanism, their empirical focus understandably has
tended to be on instances in which organizational change is engineered by initiatives ―from above.‖
The Bank‘s member states significantly influence its operations ―from above.‖ The Bank‘s
principals – the 185 member states represented on the Board – collectively approve decisions,
define the Bank‘s mandate, and oversee the management and staff‘s activities. The U.S. exercises
a great deal of influence within the Bank due to the Board‘s weighted voting system, which gives
the U.S. director the largest share of votes and the capacity to veto decisions that require a
supermajority. Though the U.S. director cannot veto a particular loan, the U.S. has historically
acted as a ―gatekeeper,‖ sometimes preventing the Bank‘s management or staff from even
proposing a loan or policy (Gwin 1997:244-245; Woods 2000:133-134; Ascher 1992:123—124).
The U.S. also has significant informal influence. Based on convention, the Bank‘s president
has always been an American citizen and therefore, according to Ascher (1992:125), likely ―to be
aware of U.S. concerns and to think like a U.S. insider.‖ But, as I document below, the Bank‘s
president has not always been subservient to Washington.
Still, U.S. officials do derive a more
subtle form of informal influence from the so-called ―Treasury effect,‖ that is, the regular formal
and informal contact that occurs between U.S. officials and the Bank‘s management and staff. This
contact enables U.S. officials to circumvent the Board entirely and address their demands directly to
the management and staff.
The importance of the U.S. capital market in the Bank‘s early years provided another formal
control mechanism with which the U.S. could potentially influence the Bank. At Bretton Woods,
6
there was a general expectation that the Bank would operate by guaranteeing securities issued by
foreign governments rather than issuing bonds of its own. However, discussions with banking and
insurance groups in 1946 made clear that private investors would prefer to hold Bank securities as
opposed to securities of foreign governments guaranteed by the Bank (Mason and Asher 1973:43—
44). Since the Bank‘s Articles require the Bank to acquire the permission of any member to
borrow in its currency, this meant the U.S. – where 85 percent of the Bank‘s bonds were sold during
its first ten years in operations – could influence Bank policy and practices by threatening to deny it
access to American capital markets (Gwin 1997:202). But the US did not use this authority until the
early 1960s and rarely employed it to place pressure on the Bank on occasions when it did. This
mechanism of influence, as Gwin (1997:245) notes, thus tended to be limited to being ―more an
implied than applied source of power.‖
Operating the Bank by issuing bonds rather than guaranteeing securities proved to be a
double-edged sword. On the one hand, issuing bonds reduced the IBRD‘s dependence on member
state contributions and thus offered a degree of autonomy from principals who sought to exert
leverage through financial control (Woods 2000:137-138). On the other hand, it introduced
another source of external influence that the Bank had to confront: private capital markets.
Although on paper the Bank initially had pledged assets of nearly $10 billion, only 20 percent of
this was paid-in subscriptions and at the time only the American and part of the Canadian
contribution was in a convertible currency. Borrowing on U.S. capital markets was thus essential to
finance the Bank‘s operations. Yet memories were long on Wall Street of the wave of defaults in
the 1930s and many state legislatures refused to allow investment in Bank bonds.
As a result, the Bank‘s second president John McCloy, along with the Bank vice president
and the U.S. director, spent much of spring 1947 on the road, giving speeches at bankers‘
conventions and lobbying various state legislatures. To ensure adequate financing for the Bank,
McCloy‘s biographer (Bird 1992:288) suggests that McCloy ―planned to run the Bank as if its
clients were private Wall Street investors and not the forty countries that had joined in the hope of
7
receiving development aid.‖ The Bank‘s second president thus concentrated on winning and
maintaining a triple-A bond rating, which meant following the conservative lending strategy
preferred by Wall Street, focusing on specific projects that generated transparent proceeds, such as
investments in capital infrastructure. The strategy ultimately proved successful, as the Bank‘s first
bond offering in July 1947 was oversubscribed. However, it would not be until 1958 that the Bank
would achieve its triple-A rating.
Evaluating the Rationalist Approach
Despite the array of formal and informal control mechanisms wielded by U.S. officials as
well as private capital markets, their influence cannot fully account for the Bank‘s early lending
practices. An examination of dynamics internal to the Bank and the authority of the management
and staff over lending decisions is also crucial. This authority was due to some of the factors
highlighted by PA theorists as well as constructivists.
At the core of the Bank‘s authority is the autonomy it derives from the process of
delegation. Many Bretton Woods delegates – including Harry Dexter White and John Maynard
Keynes - intended for the IBRD to act as a balance of payments lender of sorts (Chwieroth
forthcoming-a). While the International Monetary Fund (IMF) was to provide stabilization loans
with short repayment terms aimed at tackling short-term payments imbalances, the Bank was to be
given explicit powers to offer long-term stabilization and general purpose balance of payments
loans.2
However, the Bank‘s Articles were not entirely clear on the extent to which the Bank should
offer such loans. Inter-bureaucratic rivalries within the U.S. government and between some U.S.
officials and other Bretton Woods delegations significantly watered down the Bank‘s role as a
balance of payments lender. Instead of explicit powers, the Bank was left with an ambiguous,
complex, and somewhat conflicting mandate.
2
The Bank‘s authority to issue general purpose balance of payments loans thus overlapped with the Fund‘s authority to
disburse short-term stabilization loans, a jurisdictional issue that generated often intense friction between the two sister
institutions. The Bank often sought to maintain the fiction that such loans did not encroach on the Fund‘s authority to
facilitate adjustment; instead, such loans were aimed at preventing the disruption of development in countries faced
with the need to correct short-term payments imbalances.
8
The IBRD Articles – in a passage that came to be known as the specific project provision –
directs the purpose of lending as follows: ―Loans made or guaranteed by the Bank shall, except in
special circumstances, be for the purpose of specific projects of reconstruction and development.‖3
Another passage also indicates that the IBRD is ―to promote the long-range…maintenance of
equilibrium in the balances of payments.‖4 These provisions suggest Bank loans are to be used
primarily to finance the foreign exchange requirements of specific projects, but that the Bank can
also provide program loans under ―special circumstances.‖ Nonetheless, the Articles are unclear as
to how ―special‖ the circumstances would have to be for the Bank to provide such financing. This
is a matter that would demand interpretation.
Upon finalization, the Bretton Woods Agreement was sent to various national parliaments
for ratification, where aspects of the agreement were vigorously debated, including the extent to
which the Bank could provide program loans. The issue was addressed most extensively in the
debates in the U.S. Congress. Ultimately, it was agreed that the IBRD Articles were not sufficiently
broad to permit the disbursement of long-term stabilization and general purpose balance of
payments loans. The U.S. Congress therefore directed the U.S. governor and director of the Bank:
―to obtain promptly an official interpretation by the Bank as to its authority to make
or guarantee loans for programs of reconstruction and the reconstruction of monetary
systems, including long-term stabilization. If the Bank does not interpret its powers
to include the making or guaranteeing of such loans, the governor of the Bank
representing the United States is hereby directed to propose promptly and support an
amendment to the Articles of Agreement for the purpose of explicitly authorizing the
Bank, after consultation with the Fund, to make or guarantee such loans.‖5
The U.S. governor subsequently raised the issue at the March 1946 inaugural meeting of the
IMF and the IBRD. The IBRD governors in turn agreed to refer the issue to the Bank Board for
interpretation. Yet the Board‘s decision sidestepped the issue, failing to offer a definitive
interpretation of the Bank‘s role. A subset of Board members, sitting as the Bank Committee on
Interpretation, issued a report that was subsequently supported by the entire Board. The report
(World Bank 1946:26) concluded:
3
Article III, Section 4, Clause VII.
Article I, Section III.
5
Bretton Woods Agreements Act. Public Law 171, 79 Congress 1 session (21 July 1945), Section 12.
4
9
―Under Article III, Section 4 (vii) of the Articles of Agreement [the specific project
provision], the Bank, while primarily expected to make or guarantee loans for specific
projects of reconstruction and development, does have the authority to make or
guarantee loans for programs of economic reconstruction and the reconstruction of
monetary systems, including long-term stabilization loans, even if such loans are not
for specific projects of reconstruction and development.‖
Ultimately, however, the report offered no firm directives or guidance on how to use this authority
and left ―the Bank [management and staff] to decide whether special circumstances exist which
justify it in making or guaranteeing such loans.‖ The Board thus confirmed that the IBRD could
offer long-term stabilization and general purpose balance of payments loans, but left it open to
future interpretation as to the conditions under which such lending might be permissible. By
sidestepping the issue the Board‘s decision left the staff with considerable authority over how to
interpret this key matter.
The Bank‘s interpretative authority over this matter and others was further enhanced by
subsequent developments in the Bank‘s early history. Some of this heightened authority was due to
factors that PA theorists highlight. Yet the first, and perhaps most significant, development
directly contradicted the interests of many of the Bank‘s principals. The Bank‘s heightened
authority was more or less forced upon its principals and the U.S. in particular by the unique
circumstances surrounding McCloy‘s appointment. These unique circumstances allowed McCloy
to engineer a significant change in policies governing principal-agent relations, exercising a degree
of leverage that goes beyond what PA analysis expects.
When the Bank began its operations in 1946, some Board members – particularly U.S.
director Emilio Collado – sought to construct the Bank as an organization that would be led by a
powerful Board that initiated loans and determined policy. Collado‘s brand of ―New Deal‖ politics
clashed with the more conservative outlook of the Bank‘s first president Eugene Meyer (Bird
1992:283).
Like most of the other Bank directors, Collado believed the IBRD should be operated
as a ―development agency‖ not as a ―bank,‖ financing loans that were socially desirable but not
necessarily financially sound.
10
Meyer was generally successful in resisting Collado‘s efforts to run the Bank from the
Board. For instance, when Meyer refused to approve a loan to Chile, Collado reportedly pounded
on the Board table, demanding it be approved. Meyer then calmly refused, suggesting the Bank
was not a relief agency (Bird 1992:283; Kapur et al. 1997:78—79). However, the constant
disputes with Collado took their toll on the seventy-one-year-old Meyer, and only four months after
the Bank opened its doors he resigned. Though these disputes had prevented the Bank from
approving any loans in its first four months, it seemed that Meyer‘s resignation would leave the
Board in command. The Bank‘s first annual report (1946:5) made clear that ―matters of policy
determination‖ were the responsibility of the Board. Similarly, Davidson Sommers, an early
member of the Bank‘s staff, later recalled: ―In the beginning [there] was the Board…[They]
assumed the initiative in organizing and running the affairs of the Bank…[But] what was really
important was the very predominant position of the American Director on the Board and the role
that he played.‖ Although no loans were made until 1947, it was ―the normal thing for an applicant
for a loan to stop on his way to the Bank and get the support of the U.S. Director‖ (Cited in Gwin
1997:199).
The Truman administration, however, had difficulty finding a suitable successor for Meyer.
When asked to serve as the second Bank president, McCloy was advised by Meyer, Chester
McClain (the Bank General Counsel) and Eugene Black (vice president of Chase National Bank) to
insist on proper executive authority as a condition for accepting the position. McCloy met with
Truman administration officials and presented a list of conditions that would have to be in place
before he accepted the position. Specifically, he demanded that the U.S. not interfere in loan
negotiations nor exercise its ―gatekeeping‖ power by giving prior indication of U.S. positions on
loan applications, that he would have a free hand in personnel and administrative decisions, and that
he would have the right to select the U.S. director (Bird 1992:296; Gwin 1997:200).
U.S. officials initially rejected these conditions, but Truman overruled his advisers and told
McCloy that he would accept them (Gwin 1997:200). McCloy then secured agreement from the
11
other Board members. McCloy then negotiated a division of responsibilities among the
management, staff, and the Board, an arrangement later formalized in June 1947. Under the terms
of the arrangement, the Board was responsible for policy decisions; however, policy
recommendations were to come from the management and staff. The arrangement not only altered
the formal balance of power within the Bank, it also served to shift the informal information
environment in favour of the Bank staff, making it more difficult for the Board to monitor and
dictate lending decisions. The staff, not the Board, was charged with the responsibility of
evaluating and designing loans and it was up to the staff to keep the Board informed of the progress
of their operational work. McCloy also subsequently selected Black to replace Collado as U.S.
director, and the practice of the Bank president selecting the U.S. director remained in place until
George Woods‘s appointment as president in 1963.
The initial reactions of some Board members to these new arrangements illustrate the
extraordinary degree to which McCloy was able to minimize the extent to which member states
could shape Bank policy. In the eyes of some directors, McCloy and the Bank were close to an
agent that had run amok. The British director bluntly called McCloy‘s conditions an ―ultimatum;‖
but given the diminishing capacity of the Bank to function without a president, he and the other
directors reluctantly agreed. Reflecting the strength of the McCloy‘s position, the British director
reported back to London: ―What happens now I don‘t know but I must say that dirt is a disagreeable
diet‖ (Bird 1992:286). Similarly, later in his term when the directors lined up for a group photo with
McCloy, one director reportedly joked, ―Why don‘t the members pose with their rubber stamps in
their hands?‖ (Bird 1992:300).
Informational considerations and the need for expert knowledge further strengthened the
authority of the Bank over lending decisions.6 As the Bank‘s operations evolved, loan designs
required a degree of accumulated technical expertise and sophistication that the directors and their
staffs could not provide. Growth of Bank membership and turnover on the Board meant that
directors often did not possess the kind of institutional memory that the staff had developed over
6
For an overview of the complexity involved in crafting a Bank loan, see Baum 1970.
12
time (Ascher 1983:421). As a result, directors found it difficult to object to staff recommendations
on technical grounds (Ascher 1992; Ayres 1983:66; Woods 2000:140). As Ascher (1983:421,
emphasis added) notes: ―In practice, the Executive Directors veto a project only under extraordinary
circumstances and have virtually no opportunity to initiate the consideration of specific projects.‖
This increased autonomy from the Board has provided the staff with significant agendasetting and gatekeeping powers. The staff‘s agenda-setting powers derive from the fact that
McCloy‘s negotiations effectively relegated the Board to being primarily a decision-making body,
limited largely to approving or rejecting staff loan proposals. Generally speaking, most directors
receive detailed information about most loans at the end of the loan cycle, limiting their ability to
influence design and implementation. A number of loans are often ―automatically‖ disbursed
without Board approval due to streamlined lending procedures. Directors typically must raise an
issue related to a particular loan for it even to be brought up for discussion (Gutner 2005:29). But
due to the heavy volume of loans going to the Board each year, directors and their staffs often do
not have the ability to assess each one carefully; instead they must rely on the staff‘s judgment. The
staff‘s ability to refuse to submit a loan to the Board either because it anticipates a negative decision
or because it disapproves of some aspect of the loan provides it with gatekeeping power.7 After a
loan has been approved, the staff‘s responsibility for assessing a loan‘s implementation and for
recommending to the Board whether to disburse the next stage of the loan also give them significant
influence over lending decisions.
To be sure, the Bank‘s management and staff must always contend with the influence that
the U.S. and other member states seek to wield over Bank policy. In cases where geopolitical or
economic interests are clearly at stake, such as the decisions not to lend to Poland and
Czechoslovakia in 1947 (see below), U.S. influence is clearly instrumental. But this still leaves a
great deal unexplained. In perhaps the most comprehensive analysis of U.S. influence over the
Bank, Gwin (1997: 252, 263) concludes that ―Bank lending… [has] experienced U.S. influence,
7
Of course, the staff‘s refusal to submit a loan to the Board in anticipation of a negative decision also indicates that
member states retain a degree of control over loan decisions.
13
though less heavily than often assumed. Bank lending and U.S. policy interests have usually
converged, and the United States has not often interfered directly in specific loan decisions…Loans
made despite strong opposition by the Bank‘s largest shareholder are the exception. However, U.S.
pressure affecting the Bank‘s decision to lend to a specific country or for a specific project is also
the exception.‖8
Beyond the ―outer structural constraint‖ set by the Bank‘s principals there is thus a range of
policy choices that are shaped by the staff. PA theorists help us to understand why this might be the
case. Autonomous IO behaviour is, after all, expected. Yet PA theory is less helpful in
accounting for the processes that shape IO preference formation and change. It is not that PA
theory does not expect IOs to possess significant autonomy or discretion, but rather that it fails to
explain adequately what preferences IOs will pursue and how these preferences will change given
this autonomy and discretion. Instead, PA theorists deal with preference formation and change by
assumption.
A common assumption is that budget maximization motivates staff behaviour. Yet the
Bank‘s abandonment of program loans, which were far larger in size than project loans and thus
offered a greater potential for budget maximization, constitutes a puzzle for this view. An
examination of the role of organizational culture proves more fruitful.
Organizational Culture and Change “From Within”
Applying insights from organizational sociology, constructivists have started to explore how
organizational culture shapes staff behaviour (Barnett and Finnemore 2004; Weaver forthcoming).
Like PA theorists, constructivists see staff behaviour as partially constrained by member state
preferences, namely through an organization‘s resource dependence on powerful members and need
to conform to the normative environment these members support. But within this ―outer structural
constraint,‖ constructivists recognize IOs have considerable autonomy due in part to the legitimacy
they derive as expressions of various forms of authority.
8
See also Ascher 1992; Woods 2000.
14
Indeed, in addition to the asymmetric information considerations outlined earlier, the Bank
staff‘s autonomy was also strengthened due to their legitimacy as experts. The Bank staff were
consciously aware of the importance of this asset and sought to manage it in such a manner as to
further their autonomy.
Between 1949 and 1960, the Bank staff‘s legitimacy was reinforced by
what the Bank‘s historians (Kapur et al. 1997:457) label ―a virtual revolution in Bank staffing‖
when technical experts, mainly engineers, were heavily recruited. According to the Bank‘s
historians (Kapur et al. 1997:8), this recruitment was necessary not only to develop high-quality
project loans, but also because the staff recognized that the ―imagery‖ of expertise would help
create the appearance of depolicitization. This appearance would help the Bank to quarantine itself
against complaints that it was meddling in the affairs of its member states and to navigate the Cold
War conflict.
Constructivism offers a broad understanding of how culture bounds rationality, seeing it as
shaping and defining how individuals view organizational goals and conditioning how they interpret
initiatives for change. Initial empirical work (Weaver and Leiteritz 2005; Momani 2005) shows
how initiatives that fail to ―resonate‖ with beliefs that prevail within an organization are often
rejected rather than leading to organizational change. Organizational culture, once entrenched, is
said to be ―sticky,‖ shaping the possibilities for change.
While demonstrating that culture matters, this work tends to downplay the possibility and
efficacy of staff strategic agency. Staff are depicted as overly socialized creatures of habit that are
incapable of looking beyond a particular world view and guided solely by the ―logic of
appropriateness.‖ An unduly static understanding of staff behaviour thus plagues work on
organizational culture.
However, if we turn our attention to internal IO debates we can inject a greater degree of
agency and dynamism into otherwise structuralist and static constructivist accounts. Staff can
become socialized into their cultural environments, but they also retain the capacity to strategically
15
reconstruct that culture. In other words, they can act as internal entrepreneurs or ―insider
proponents‖ for normative change (Finnemore and Sikkink 1998; Acharya 2004).
Though appearing under various names, the concept of ―norm entrepreneurs‖ has
distinguished intellectual history, dating back to Machiavelli, on which I build.9 Using some form
of persuasion, social pressure, material incentives, or position of leadership, these actors engage in a
process of ―strategic social construction‖ in which they manoeuvre through the organizational
bureaucracy promoting new beliefs, which can potentially alter an organization‘s culture and
practices.
Thus, as Barnett (1998:9) suggests, understanding the behaviour of staff requires a
―blending of homo economicus and homo sociologicus.‖
Traditional accounts of IO change generally focus on norm entrepreneurs that operate ―from
above‖ or ―from below,‖ unduly neglecting how change can come ―from within.‖ Though staff
tend not to advocate policies that their principals oppose, their behaviour goes beyond being simply
reactive actors who respond to external initiatives. As a growing number of empirical studies
suggest, staff can also be proactive actors at the forefront of efforts to promote organizational
change (Bebbington et al. 2004, 2006; Fox 1998; Kardam 1993).
I extend and refine the nascent literature on internal IO norm entrepreneurs by identifying a
critical internal mechanism through which they surface and a vital condition that helps them
facilitate change. The realignment of personnel and institutional configurations stand out as critical
processes. Personnel realignment, which is driven by administrative recruitment, occurs when staff
wedded to particular beliefs are replaced or supplanted by new staff sharing alternative beliefs.
Sociologically inflected studies of organizations often allude to how recruitment patterns shape staff
beliefs (Dimaggio and Powell 1983; Ascher 1983; Miller-Adams 1999; Barnett and Finnemore
2004). Yet these studies typically treat recruitment as mechanism that works to maintain
organizational culture. Missing is an appreciation and careful analysis of how recruitment can also
work to change it.
9
For an overview, see Campbell 2002.
16
Personnel configurations and recruitment procedures matter because the type of individuals
and the credentials demanded can exert a significant impact on an institution‘s views (Dimaggio
and Powell 1983; Hall 1989; Chwieroth 2007; 2008). In particular, if an organization – such as the
Bank – recruits from a range of professions, then its practices will be highly susceptible to
differences that exist across these professions; and these differences will be critical for
understanding the evolution of organizational behaviour. Subcultures are likely to form as
individuals from different professions, and thus sharing different interpretations and standards of
behaviour, are recruited into an organization. New recruits from the same profession will tend to
bring with them beliefs that often conflict with those of existing staff as well as other new recruits
from different professions. These subcultures in turn offer a fertile source for norm entrepreneurs.
The success of norm entrepreneurs often depends on their possession of a sufficient amount
of discursive influence within an organization that enables them to outperform advocates of
alternative views. This influence is in large part conditional upon their position within an
organization.
Internal norm entrepreneurs are likely to be more successful if they operate within
institutional configurations that provide them with sufficient authority to prevail upon others to
accept their arguments.10
Institutional configurations shape the relative leverage that actors have over the policy
process and agenda. This leverage could be informal or formal. Informally, though they may not
be at the apex of formal institutional structures, staff with relatively long careers within an
organization or profession from which it recruits are often constituted as ―an authority‖ on
particular issues. Other staff with comparatively less technical knowledge, experience, and
institutional memory will therefore be wary of challenging their diagnoses and prescriptions, and
are likely to defer to their arguments.
The formal organization of decision-making and the positions that individuals occupy within
the bureaucratic matrix may also give norm entrepreneurs certain advantages that place them ―in
10
This argument is consistent with constructivist arguments about the need for an organizational platform, see
Finnemore and Sikkink 1998.
17
authority.‖ By occupying strategic locations with an organization, internal advocates can obtain a
significant degree of leverage over the policy process and agenda (Hall 1989; 1993). From these
locations norm entrepreneurs are better positioned to gain access to and present their views to
management, to launch new initiatives, to acquire and control information and other resources, and
in some instances to serve as a ―veto point.‖ The efforts of norm entrepreneurs are also often
supported by realignments of institutional arrangements and policy-making procedures that
strengthen their position within the organizational bureaucracy.
Changing Lending Practices “From Within”
In the years immediately following the war, many countries – especially in Western Europe
– faced severe balance of payments deficits. Despite the dollar shortage and the payments deficits
facing Europe, the IMF had a relatively minor financing role. The IMF‘s capacity to lend was
paralyzed by debates that persisted between the U.S. and other countries over whether IMF
resources should be disbursed unconditionally (Chwieroth forthcoming-a).
Although the operations of the Bank began with a rather rocky start, one of Meyer‘s
principal accomplishments was to establish the Bank‘s internal decision-making structure. The
Bank was to be managed by a president and vice president, with equal authority over lending
operations given to the Loan and Research Departments (World Bank 1946a:8; 1946b:21).
Initially, the Loan Department was staffed mainly with bankers and lawyers, whereas the Research
Department (renamed the Economics Department in 1948) was manned primarily by economists;
interdepartmental conflict was to be resolved in the Staff Loan Committee (later renamed the Loan
Committee), made up of the principal department heads and chaired by the vice president (Kapur et
al. 1997:456).
Relations between the Loan and Research Departments were never harmonious, and early in
the history of the Bank there was an important ―battle of ideas‖ about how its resources should be
used. Paul Rosenstein-Rodan, the assistant director of the Research Department and a leading
development economist of the era, was the principal voice of the economists. From an operational
18
perspective, Rosenstein-Rodan criticized the disbursement of project loans, viewing them as
plagued by a ―fungibility problem.‖ Rosenstein-Rodan worried that the Bank risked financing
investments that a government might have undertaken on its own, or, potentially worse, providing
financing that could all to easily be siphoned off for purposes that ran contrary to Bank interests.
Since monitoring the use of Bank financing for specific projects would prove difficult, RosensteinRodan argued that the Bank should devote its resources to providing program loans for general
purposes.
From a development perspective, Rosenstein-Rodan doubted whether a series of specific
projects would prove effective. Only with a ―big push‖ in investment within the context of a
program, Rosenstein-Rodan argued, could an economy transform itself from underdeveloped to
developing (Rosenstein-Rodan 1944; Kapur et al. 1997:128; Oliver 1975:272). It was idle,
Rosenstein-Rodan claimed, for the Bank to concern itself with borrower creditworthiness and
specific projects; rather, it should calculate the financing necessary to sustain a desired growth rate
and make massive loans available on a continuing basis. Although the bankers and lawyers in the
Loan Department shared the technocratic bent of the economists, their professional norms led them
to view creditworthiness as the key factor determining a borrower‘s eligibility. For these staff
members the IBRD was to be a ―bank‖ rather than a ―development agency.‖ Rosenstein-Rodan‘s
advice therefore ―seemed like nonsense‖ (Oliver 1975:272).
This battle of ideas took on added significance because of personnel problems at the top of
the Bank hierarchy. McCloy became the second Bank president in March 1947 and immediately
faced the issue of what the role of the Bank should be in Europe. By April 1947, six countries had
submitted applications for Bank loans, and still another was to received in August. Though the IMF
had been established to provide balance of payments financing, its capacity to lend was paralyzed
by the conditionality debates. Moreover, though Marshall Plan aid would later marginalize the
Bank‘s role in Europe, in 1947 the shape and timing of U.S. aid was still in the process of
formulation.
19
But the Bank was still very much in its infancy, and its organizational culture had yet to be
established and was still under debate. As a result, no one within the Bank really knew where to
begin, what types of questions to ask, and what sort of investigation to undertake in response to the
loan applications. Though recognizing that Europe needed a massive disbursement of aid, McCloy
was uncertain as to whether it should be in the form of project or program loans.
McCloy also faced the delicate task of managing external pressures. U.S. officials initially
pressured McCloy to lend exclusively to Western Europe in exchange for a larger U.S. financial
contribution to the Bank. McCloy refused, viewing the proposal as likely to transform the Bank
into an American organization rather than a multilateral organization (Kapur et al. 1997:76).
U.S.
officials then encouraged McCloy to lend to France, one of the countries that had applied for a loan,
in order to bolster the government vis-à-vis the Communist Party (which had a minor position in
government and was seen as likely to increase its mandate in the next elections) (Bird 1992:290).
At the same time, McCloy faced pressures from private capital markets. As suggested,
McCloy and others believed that project loans were necessary to strengthen the Bank‘s
creditworthiness.
There was also the need to manage the implied, but not applied, pressure that
could potentially be exerted on the Bank should U.S. officials decide to deny it access to American
capital markets.
These pressures likely exerted a significant influence on Bank lending practices. U.S.
officials reinforced McCloy‘s inclination to lend to France and exerted a significant degree of
influence on his decision not to lend to Poland and Czechoslovakia, which had also applied for
loans (Gwin 1997, 253—254).
Private capital market constraints also likely shaped early Bank
lending practices. Despite criticism from some member states who envisioned the Bank taking a
larger and lead role in financing post-war reconstruction and development, McCloy was generally
committed to a conservative lending strategy that focused on financing capital infrastructure and
offered little lending to social sectors. Due to constraints imposed by the need to sell bonds to
private capital markets, McCloy insisted that the Bank could not substitute, even in the interim, for
20
the massive American aid that would be needed (Gwin 1997:202; McCloy 1949; Bird 1992:292293). But, these constraints cannot fully explain the Bank‘s early lending profile.
There is little evidence that U.S. officials or private capital market constraints were decisive
in shaping McCloy‘s decision about what form the Bank‘s loans should take. U.S. officials were
actively involved in shaping the policy conditions under which the French loan would be made
available. Yet they were indifferent about the form the lending was to take.
Attributing Bank lending practices to private capital market constraints is also somewhat
problematic. According to his biographer (Bird 1992:290), McCloy initially felt that
recommending program loans to the Board would threaten the Bank‘s standing with private capital
markets, as ―[It] was exactly the kind of loan that ordinary commercial bankers would avoid, the
kind of loan that had gone into default in the 1930s.‖ But program loans were precisely what
Rosenstein-Rodan and other Bank economists argued that McCloy should recommend. It was in
this environment that Rosenstein-Rodan and his economist colleagues convinced McCloy to go
against private capital market constraints and recommend program rather than project loans to the
Board. As McCloy‘s biographer (Bird 1992:291) notes: ―[T]hough he had not initially been
prepared to endorse such lending, McCloy was quickly convinced by the Bank‘s senior economists
that Europe‘s war-torn economies needed such balance-of-payments financing.‖11 The economists
were also successful in persuading McCloy to recommend program loans for four other countries,
all of which were approved by 1948.
These loans, however, marked the highpoint of the influence of economists on early Bank
lending practices. As late as 1949, the Bank‘s annual report (1949:9) still contained staff economist
warnings about the fungibility problem associated with project loans. These warnings, however,
were subsequently removed due to the emergence of an alternative perspective that prioritized
projects as the norm. Over the next few years the economists‘ views would become increasingly
marginalized within the Bank and their authority markedly diminished.
11
See also Mason and Asher 1973:51-52.
21
The bankers and the lawyers in the Loan Department had generally objected to Board; and
their arguments were received sympathetically by Vice President Robert Garner, to whom McCloy
delegated considerable responsibility. As vice president, Garner chaired the Staff Loan Committee
and, as Oliver (1975:239) notes, was ―probably more responsible than any other single person for
the evolution of the Bank through the mid-fifties.‖ Garner did not understand and had little
patience for the Bank‘s economists. Oliver indicates that Garner would blue pencil terms like
―capital-output ratio‖ out of reports, calling them ―economeeze.‖ Garner also generally opposed
program loans and felt the Bank should confine itself to financing specific capital infrastructure
(Oliver 1975:239-249 Kraske 1996:55, 90—91).
To promote the norm of projects within the Bank, the bankers and lawyers within the Loan
Department engineered a series of institutional and personnel realignments that facilitated the
construction of a ―project-oriented‖ culture. The first institutional realignment occurred in 1948
when the Stanley Hoare, assistant director of the Loan Department, persuaded Garner to remove the
Economics Department from having any responsibility over Bank lending operations (Oliver
1975:273). In the aftermath of this institutional realignment, the views of the bankers and lawyers
began to prevail, and Bank lending became more oriented toward project loans (Mason and Asher
1973:458-461; Oliver 1975:25; World Bank 1949:7—8). To help prevent the disbursement of
program loans, the Loan Department would typically delay the approval of applications until the
specific purposes of a proposed loan could be properly ascertained (Kapur et al. 1997:105—106;
Mason and Asher 1973:155; Oliver 1975:246).
The emergence of the project-oriented culture was reinforced by the realignment of Bank
personnel that occurred between 1949 and 1960. Engineers were recruited on a massive scale,
altering the personnel profile of the Bank in a ―pro-projects direction‖ (Kapur et al. 1997:458). As
engineers, these new staff tended to see development as being a sequence of new physical structures
or projects that had to be put in place and thus favoured project loans (Sadove 1967; World Bank
1957:14; Mosley et al. 1991:29).
22
In 1952, internal norm entrepreneurs successfully engineered another institutional
realignment that further strengthened the development a project-oriented culture. This institutional
realignment was designed by a staff committee chaired by the Loan Department director (Mason
and Asher 1973:75). According to one Bank historian (Oliver 1975:173), ―The primary reason for
the reorganization of 1952 was to give the loan or operations departments more power, and the
research or economics department less power.‖ The most powerful department in the Bank became
the new project-oriented Technical Operations Department (TOD, later renamed the Projects
Department), while the Economics Department lost its departmental status, becoming the ―small
and underfunded‖ Economics Staff (Mason and Asher 1973:467; Oliver 1975:100, 158; Kapur et al.
1997:129-130; 458-461). New area departments were also created to manage the Bank‘s relations
with borrowers. Although some economists, such as Rosenstein-Rodan, left the Bank, others joined
the other departments; those that remained tended to adopt the project-oriented culture.
The creation of the TOD facilitated the recruitment of additional engineers and project
specialists, thus further strengthening their views within the Bank. The TOD and its director as
chair of the Loan Committee emerged from the realignment with firm control over Bank operations.
As the Bank increasingly specialized in project work, it developed operational routines – known as
the project cycle - to guide its lending.12 These operational routines served to marginalize the
impact that the newly created area departments could have on crafting Bank loans.
By prioritizing the disbursal and implementation of projects, the project cycle had the effect
of minimizing the degree to which the area departments could raise concerns about the policy
environment of borrowers (Please 1984:27—28). Traditionally, staff responsible for the Bank‘s
relations with borrowers had focused on improving the quality of the policy environment in which
Bank loans were disbursed. In fact, staff were quite open about their efforts to influence the
policies of borrowers.13
12
13
See Baum 1970.
See, for instance, World Bank 1955:35.
23
McCloy was also convinced of the desirability of the Bank‘s early efforts to encourage
policy reform. In a spring 1949 article in Foreign Affairs, McCloy (1948/1949:551) noted: ―Its
advice and encouragement to various of its members have led to the adoption of economic and
financial measures and administrative reforms which may well prove to be at least equally as
beneficial as the Bank‘s financial help.‖ But McCloy left the Bank in July 1949 to serve as High
Commissioner in Germany, and by the 1960s the Bank‘s annual reports no longer contained
remarks about policy reform, these having been replaced by a new emphasis on the Bank‘s ability
to initiate and help develop specific projects (Ascher 1992:130).
Despite the ascendance of the project-oriented culture, Bank economists did on occasion
manage to persuade Eugene Black, who was McCloy‘s choice to replace him when he left the Bank,
to recommend the disbursement of program loans. Their arguments initially gained currency in part
due to mounting concerns within the Bank over a decrease in lending in the third fiscal year of
operations. The Marshall Plan and Point Four Program had threatened to make the Bank irrelevant.
While project-oriented staff pointed to a shortage of feasible investments and creditworthy
borrowers, the economists argued that the Bank‘s project-oriented approach was to blame (Mason
and Asher 1973:268-269). On this debate, McCloy tended to support the views advanced by the
project-oriented staff (McCloy 1948/149:554; Gwin 1997:204).
Black, on the other hand, seems to have been convinced by some of the arguments the
economists offered, particularly after Garner left in 1956 to lead the Bank‘s new International
Finance Corporation (Oliver 1975:240, 273; Kapur et al. 1997:134). After Garner‘s departure, three
new vice presidents were promoted within the Bank, one of whom, J. Burke Knapp, was the former
third-ranking official within the Economics Department. In the reorganization Knapp was also
appointed to chair the Staff Loan Committee. Black showed himself to be more willing to listen to
the arguments the economists offered.
Between 1950 and 1957 the Board approved program loans for five countries. During this
period, the Board tended to greet the approval of program loans with great enthusiasm (Mason and
24
Asher 1973:271-273). But the Board generally lacked the capacity to initiate such loans due to the
arrangements McCloy had negotiated. Even the U.S. exerted little significant pressure on the Bank
president to initiate loans during this period (Mikesell 1954:200). To be sure, there were always a
few directors who were sceptical of program lending. However, none of these program loans
triggered any significant debate on the Board about their legitimacy (Mason and Asher 1973:270,
272).
Despite these setbacks, the project-oriented culture generally prevailed. As opposed to the
period prior to the institutional and personnel realignments engineered by the project-oriented staff,
recommendations for program loans tended to be rare and met by opposition from the TOD (Mason
and Asher 1973:271, 468; Oliver 1975:291). Whereas program loans constituted 73.2 percent of
approved financing from 1946 – 1950, these loans constituted only seven percent of approved
financing from 1951 – 1957.14 Reflecting the triumph of the project-oriented culture, the Bank‘s
tenth annual report (World Bank 1955:32), looking back on its first decade of operations, notes:
―The most typical pattern of lending has been and will continue to be a series of single loans, made
over a period of time, to finance imports for a variety of single projects.‖
The end result of this series of institutional and personnel alignments and battle of ideas was
the construction of a project-oriented culture and a set of institutional arrangements and policymaking procedures that lodged the authority to initiate loans largely in the TOD. The construction
of this project-oriented culture played a crucial part in leading the Bank to abandon program loans.
The ―special circumstances‖ language in the specific project provision was ambiguous and complex
and left to the Bank management and staff to interpret for themselves. This gave the Bank
tremendous flexibility in shaping its early lending practices. The Bank‘s project-oriented culture
was not permissive of program loans and as a result, as Dell (1984:165) observes, ―the Bank‘s clear
authority in this regard was allowed to atrophy.‖
In part due to the dominance of the engineers and the project-oriented culture, the Bank, as
its historians (Mason and Asher 1973:468) note, ―was slow to break away from its early devotion to
14
Author calculations based on World Bank, various years.
25
capital infrastructure.‖15 To be sure, during the 1960s the Bank broadened its focus, providing
loans for education, agriculture, and water supply and sanitation. Yet the Bank‘s culture was still
oriented toward projects. According to the Bank‘s historians (Kapur et al. 1997:8, 9), ―by the end
of the 1950s the culture of the Bank had become project-led‖ and ―this project culture…had a
marked effect on the Bank‘s history.‖ Indeed, between 1957 and 1966 the Bank failed to approve
any program loans.
Some may be tempted to attribute the absence of program lending to private capital market
constraints. These constraints were likely to have been the most severe when Bank management
sought to gain the confidence of investors prior to the achievement of a triple-A bond rating. These
constraints then likely diminished once investor confidence was achieved. Therefore, if private
capital market constraints alone can explain the Bank‘s early lending practices, then one would
expect a higher proportion of project loans prior to the Bank‘s achievement of a triple-A bond rating
than afterwards.
Yet this was not the case. The Bank achieved a triple-A bond rating in 1958 and has since
maintained it. As discussed, before 1958 some Bank staff showed themselves to be quite willing –
particularly prior to the institutional and personnel realignments engineered by project-oriented staff
– to recommend program loans, despite private capital market constraints. According to Gwin,
after the achievement of a triple-A bond rating the Bank‘s management and staff felt ―less
constrained‖ by the attitudes of private capital markets (Gwin 1997:203). In addition, by the mid1950s the Bank had been able to broaden its investor base to non-American markets such that by
the late 1960s more than half its bonds were being purchased by investors outside the United States
(Gwin 1997:203).
If private capital market constraints alone can fully explain the Bank‘s lending practices,
then one would expect the staff to have recommended at least some programs loans after the
achievement of the triple-A bond rating. But the staff failed to do so until the late 1960s. This point
here is not to deny that private capital market constraints were a crucial factor shaping Bank
15
See also Oliver 1975:291.
26
lending. Rather, the point is to suggest that private capital market constraints alone cannot fully
explain the Bank‘s early lending practices. It should also be noted that the failure to recommend
program loans cannot be fully attributed to an absence of demand, as developing countries had long
been making arguments in favour of the Bank providing program loans (Kapur et al. 1997:155—
159, 168). A fuller understanding of the Bank‘s early lending practices thus necessitates a focus on
the Bank‘s intra-organizational dynamics and how the staff understood the legitimacy of certain
types of loans.
Conclusion
Some factors highlighted by PA theorists – such as ambiguous, complex, and conflicting
mandates and asymmetric information – were critical in providing the Bank sufficient autonomy.
Yet PA theory is less useful in helping us to understand how the Bank went about using this
autonomy to develop its lending practices. Internal processes and collectively shared beliefs
appear largely responsible for shaping the trajectory of the Bank‘s early lending practices.
The
Bank‘s administrative recruitment, which drew from a range of professions, was critical in
facilitating the creation of subcultures and creating an environment that was ripe for a battle of
ideas.
In the battle of ideas that ensued, success was often conditional upon one‘s position within
the Bank. Rosenstein-Rodan‘s informal status as ―an authority‖ certainly helped him overcome the
formal institutional disadvantage of serving as the second highest-ranking official in the Economics
Department below a banker who had little training in economics (Kapur et al. 1997:77). In terms of
formal positional advantages, the 1948 and 1952 institutional realignments clearly strengthened
project-oriented staff. Finally, consistent with constructivist arguments, norm entrepreneurship also
seems to be facilitated by the extent to which arguments and policy prescriptions resonate with
existing beliefs. For instance, Hoare‘s capacity to convince Garner to remove the Economics
Department from having responsibility for loans was significantly aided by the fit between his
arguments and Garner‘s general scepticism of economics.
27
Several conclusions follow from this analysis. Theoretically, it implies that greater attention
needs to be devoted to internal processes that shape IO behaviour. PA theory usefully highlights
some of the factors giving rise to IO autonomy, but falls short in helping us to understand what IOs
do with this autonomy. The constructivist turn to organizational culture usefully rectifies this
deficiency, but often at the expense of offering overly structuralist and static depictions of staff
behaviour. By shining light on some sources of IO preferences and theorizing some of the
processes through which staff members can strategically affect change ―from within,‖ this analysis
offers a useful step in building a more synthetic and holistic view of organizational change.16
Focusing on intra-organizational dynamics offers a powerful way for understanding the
potential for organizational change. Debates among staff have been often critical in shaping the
trajectory of organizational practices throughout the Bank‘s history. The early staff debates over
program and project loans resurfaced often in the organization‘s history, with the battle of ideas
among the staff often decisively shaping practice. For instance, advocacy by program loan
proponents – such as Irving Friedman and Ernest Stern – was vital in shaping the decision to
disburse program loans to India in the late 1960s as well as structural adjustment loans in the 1980s
(Chwieroth 2005). The staff‘s role in precipitating Paul Wolfowitz‘s recent resignation as Bank
president also attests to the continuing importance of focusing on intra-organizational dynamics.
The Bank‘s recruitment patterns also remain relevant for understanding the potential for
organizational debate and change. Today the Bank recruits from an even broader range of
professional backgrounds than it did in its early years, which continues to give rise to subcultures
and to engender vigorous staff debate. One can certainly see internal debates over resettlement
policies (Fox 1998) and social capital (Bebbington et al. 2004; 2006), which tended to pit staff
members from different professional backgrounds against one another, in this manner.
One final conclusion concerns how IOs behave in their infancy, which the existing literature
has not yet adequately addressed. The Bank‘s early years suggest that an IO‘s infancy may often be
16
See also Nielson, Tierney, and Weaver 2006.
28
a period when things are more or less up for grabs.17
It is a period when both principal-agent
relationships and organizational culture are underdeveloped, which in turn seems to impact
organizational performance and the potential for reform. Yet due attention has not been given to
the important fluidity that can characterize an IO‘s infancy. Instead, PA theorists (Hawkins et al.
2006a:19-20; see also Ikenberry 2001) often depict great powers as simply locking in their
preferences in new IOs in a way that does not reveal much about the messiness that typically
surrounds any organization as it opens for business for the first time. Constructivists, on the other
hand, tend to take organizational culture as a given without exploring how it initially develops.
Neither approach is particularly fruitful because we may miss out on how this fluidity shapes
principal-agent relationships, culture, and organizational behaviour and performance. A profitable
area of future research would be to focus on how an IO‘s ―life cycle‖ shapes organization behaviour
and performance and the potential for reform.
17
The infancy of the IMF suggests a similar conclusion.
29
References
Acharya, Amitav. 2004. ―How Norms Spreads: Whose Ideas Matter? Norm Localization and
Institutional Change in Asian Regionalism,‖ International Organization 58:239--278.
Ascher, William. 1983. ―New Development Approaches and the Adaptability of International
Agencies: The Case of the World Bank,‖ International Organization 37:415-439.
-------. 1992. ―The World Bank and U.S. Control,‖ in M. Karns and K. Mingst (eds) The United
States and Multilateral Institutions: Patterns of Changing Instrumentality and Influence.
Boston: Unwin Hyman.
Ayres, Robert L. 1983. Banking on the Poor: The World Bank and World Poverty. Cambridge, MA:
MIT Press.
Baum, Warren E. 1970. ―The Project Cycle,‖ Finance and Development 7: 2-13.
Barnett, Michael. 1998. Dialogues in Arab Politics: Negotiations in Regional Order. New York:
Columbia University Press.
-------, and Martha Finnemore. 2004. Rules for the World: International Organizations in Global
Politics. Ithaca, NY: Cornell University Press.
-------, and Liv Coleman. 2005. ―Designing Police: Interpol and the Study of Change in
International Organizations,‖ International Studies Quarterly 49: 593-619.
Bebbington, Anthony, Scott Guggenheim, Elizabeth Olson, and Michael Woolcock. 2004.
―Exploring Social Capital Debates at the World Bank,‖ Journal of Development Studies 40: 33-64.
Bebbington, Anthony, Michael Woolcock, Scott Guggenheim, and Elizabeth Olson (eds.). 2006.
The Search for Empowerment: Social Capital as Idea and Practice at the World Bank.
Bloomfield, CT: Kumarian Press.
Bird, Kai. 1992. The Chairman: John J. McCloy and the Making of the American Establishment.
New York: Simon and Schuster.
Campbell, John. 2002. ―Ideas, Politics, and Public Policy,‖ Annual Review of Sociology 28:21-38.
Chwieroth, Jeffrey. 2005. ―‘Shopping with the Sisters‘: Principals, Organizational Culture, and
Institutional Strategies for Balance-of-Payments Financing.‖ Paper presented at the annual
meeting of the American Political Science Association, Washington, D.C., 1 – 4 September.
-------. 2007. ―Testing and Measuring the Role of Ideas: The Case of Neoliberalism in the
International Monetary Fund,‖ International Studies Quarterly 51:5-30.
-------. 2008. ―Norm Adoption ―From Within:‖ The International Monetary Fund‘s Approach to
Capital Account Liberalization,‖ International Studies Quarterly 52:129-158.
--------. Forthcoming-a. ―The Bank, the Fund, and the Provision of International Liquidity, 1942—
1957‖ in David Andrews (ed) Orderly Change: The Evolution of Bretton Woods. Ithaca, NY:
Cornell University Press.
De Vries, Margaret Garritsen. 1985. International Monetary Fund 1966—1971: System Under
Stress, Volume 1. Washington, DC: IMF.
Dell, Sidney. 1984. ―A Note on Stabilization and the World Bank,‖ World Development 12:165167
DiMaggio, P.J. and W.W. Powell. 1983. ―The Iron Cage Revisited: Institutionalized Isomorphism
and Collective Rationality in Organizational Fields,‖ American Sociological Review 48:147160.
30
Finnemore, Martha, and Kathryn Sikkink. 1998. International Norm Dynamics and Political
Change, International Organization 52:887-918.
Gould, Erica. 2003. ―Money Talks: Supplementary Financiers and International Monetary Fund
Conditionality,‖ International Organization, 57:551-586.
Gutner, Tamar. 2005. ―Explaining the Gaps between Mandate and Performance: Agency Theory
and World Bank Environmental Reform,‖ Global Environmental Politics 5:10-37.
Gwin, Catherine. 1997. ―U.S. Relations with the World Bank, 1945 – 1992,‖ in Devesh Kapur,
John P. Lewis, and Richard Webb (eds) The World Bank: Its First Half Century, Volume 2.
Washington, DC: Brookings, pp. 195—274.
Hall, Peter. 1989. The Power of Economic Ideas. Princeton: Princeton University Press.
-------. 1993. ―Policy Paradigms, Social Learning, and the State: The Case of Economic
Policymaking in Britain,‖ Comparative Politics 25: 275-296.
Hawkins, Darren G., David A. Lake, Daniel L. Nielson, and Michael J. Tierney. 2006a.
―Delegation under Anarchy: States, International Organizations, and Principal-Agent Theory‖
in Darren G. Hawkins, David A. Lake, Daniel L. Nielson, and Michael J. Tierney (eds)
Delegation and Agency in International Organizations. Cambridge: Cambridge University
Press.
Hawkins, Darren G., David A. Lake, Daniel L. Nielson, and Michael J. Tierney (eds). 2006b.
Delegation and Agency in International Organizations Cambridge: Cambridge University
Press.
Ikenberry, G. John. 2001. After Victory: Institutions, Strategic Restraint, and the Rebuilding of
Order after Major Wars. Princeton: Princeton University Press.
Kapur, Devesh, John P. Lewis, and Richard Webb (eds). 1997. The World Bank: Its First Half
Century, Volume 1. Washington, DC: Brookings.
Kardam, Nurket. 1993. ―Development Approaches and the Role of Policy Advocacy: The Case of
the World Bank,‖ World Development 21: 1773-1786.
Kraske, Jochen. 1996. Bankers with a Mission: The Presidents of the World Bank, 1946—1991.
Oxford: Oxford University Press.
Lake, David A., and Matthew D. McCubbins. 2006. ―The Logic of Delegation to International
Organizations,‖ in Darren G. Hawkins, David A. Lake, Daniel L. Nielson, and Michael J.
Tierney (eds) Delegation and Agency in International Organizations. Cambridge: Cambridge
University Press.
McCloy, John J. 1949. ―The Lesson of the World Bank,‖ Foreign Affairs 27:551—560.
Miksell, Raymond F. 1952. United States Economic Policy and International Relations. New
York: McGraw-Hill.
Miller-Adams, Michelle. 1999. The World Bank: New Agendas in a Changing World. New York:
Routledge.
Momani, Bessma. 2005. ―Limits of Streamlining Fund Conditionality: IMF‘s Organizational
Culture,‖ Journal of International Relations and Development, 8:142-163.
Mosley, Paul, Jane Harrigan, and John Toye. 1991. Aid and Power: The World Bank and Policybased Lending, Volume 1. London: Routledge.
Nielson, Daniel, and Michael J. Tierney. 2003. ―Delegation to International Organizations:
Agency Theory and World Bank Environmental Reform,‖ International Organization, 57:241276.
31
Nielson, Daniel L., Michael J. Tierney, and Catherine E. Weaver. 2006. ―Bridging the RationalistConstructivist Divide: Re-engineering the Culture of the World Bank,‖ Journal of International
Relations and Development, 9: 107-139.
Niskanen, William. 1971. Bureaucracy and Representative Government. Chicago: Aldine.
Oliver, Robert. 1975. International Economic Cooperation and the World Bank. London:
Macmillan.
Please, Stanley. 1984. The Hobbled Giant: Essays on the World Bank. Boulder, CO: Westview.
Rosenstein-Rodan, Paul. 1944. ―Problems of Industrialization of Eastern and South-Eastern
Europe,‖ Economic Journal.
Pollack, Mark A. 2003. The Engines of European Integration: Delegation, Agency, and AgendaSetting in the EU. Oxford: Oxford University Press.
Sadove, Robert. 1967. ―Economists, Engineers, and Development,‖ Finance and Development 4:
125-132.
Weaver, Catherine E. forthcoming. Hypocrisy Trap: The Rhetoric, Reality, and Reform of the
World Bank. Princeton: Princeton University Press.
-------, and Ralf J. Leiteritz. 2005. ―‘Our Poverty is a World Fully of Dreams:‘ Reforming the World
Bank,‖ Global Governance 11:369-388.
Woods, Ngaire. 2000. ―The Challenges of Multilateralism and Governance,‖ in Christopher L.
Gilbert and David Vines (eds) The World Bank: Structure and Policies. Cambridge, UK; New
York: Cambridge University Press.
-------. 2006. The Globalizers: The IMF, the World Bank and their Borrowers. Ithaca, NY:
Cornell University Press.
World Bank. Various years. Annual Report. Washington, DC: World Bank.
32